Resveratrol: Isomeric Molar Absorptivities and Stability - Journal of

May 16, 1996 - Juan F. Moreno-Labanda, Ricardo Mallavia, Laura Pérez-Fons, Victoria Lizama, Domingo Saura, and Vicente Micol .... Concentrations of r...
74 downloads 11 Views 144KB Size
J. Agric. Food Chem. 1996, 44, 1253−1257

1253

Resveratrol: Isomeric Molar Absorptivities and Stability Brent C. Trela and Andrew L. Waterhouse* Department of Viticulture and Enology, University of California, Davis, California 95616

Resveratrol has attracted interest as a wine constituent that may reduce heart disease. Published data on the molar absorptivity and chemical stability of cis- and trans-resveratrol have varied greatly. Accurate values for UV absorbance for trans-resveratrol [UV λmax (EtOH) nm () 308 (30 000)] and cis-resveratrol [UV λmax (EtOH) nm () 288 (12 600)] were determined and are used to improve chromatographic quantitation methods. Trials conducted under a variety of commonly encountered laboratory conditions show that trans-resveratrol is stable for months, except in high-pH buffers, when protected from light. cis-Resveratrol was stable only near pH neutrality when completely protected from light. Keywords: Antioxidant; phenolic; wine; grape; stilbene; isomer; UV spectroscopy; absorptivity INTRODUCTION

The non-flavonoid phytoalexin trans-resveratrol (trans3,5,4′-trihydroxystilbene) and its various derivatives are widely distributed in gymnosperms and dicotyledons (Gorham, 1980) including groundnuts (Ingham, 1976; Schoeppner et al., 1984). trans-Resveratrol was discovered in the grape vine, Vitis vinifera, and was observed to occur there as a response to fungal infection or injury in 1976 (Langcake and Pryce, 1976). Later studies revealed trans-resveratrol in grape skins (Creasy and Coffee, 1988; Jeandet et al., 1991, 1995a; Roggero and Garcia-Parilla, 1995) and more recently in wines (Siemann and Creasy, 1992; Lamuela-Ravento´s and Waterhouse, 1993; Jeandet et al., 1993, 1995b,c; Mattivi, 1993; Roggero and Archie, 1994; McMurtrey et al., 1994; Pezet et al., 1994; Goldberg et al., 1994, 1995a; Vrhovsek et al., 1995). trans-Resveratrol has been suggested as a cause for reduced CHD mortality in wine drinkers and numerous studies describe potential mechanisms by which trans-resveratrol could reduce heart disease. trans-Resveratrol has been shown to inhibit platelet aggregation (Chung et al., 1992; Kimura et al., 1985; Pace-Asciak et al., 1995; Bertelli et al., 1995), protect the liver from lipid peroxidation (Kimura et al., 1983), and inhibit low-density lipoprotein oxidation (Frankel et al., 1993). These studies of health effects were, in part, stimulated by previous reports of the physiological activity of resveratrol-containing polygonaceous plants such as Polygonum cuspidatum used in China and Japan. These plants were used as herbal folk remedies, such as Kojo-kon, for the treatments of various fungal, bacterial, inflammatory, and lipid atherosclerosis ailments (Nonomura et al., 1963; Kimura et al., 1983, 1985). Resveratrol occurs in the cis and trans isomeric forms (Figures 1 and 2). In Vitaceae fungal infection or UV light stimulates the production of stilbene synthase and catalyzes the reaction of 4-hydroxycinnamoyl-CoA and malonyl-CoA to produce trans-resveratrol (Fritzemeier and Kindl, 1981). In the grape berry, trans-resveratrol production is stimulated by UV light exposure, fungal infection, or injury (Creasy and Coffee, 1988; Jeandet et al., 1991, 1995a). cis-Resveratrol has not been reported in Vitis vinifera; however, it has been shown * Author to whom correspondence should be addressed (e-mail [email protected]). S0021-8561(95)00457-2 CCC: $12.00

Figure 1. trans-Resveratrol.

Figure 2. cis-Resveratrol.

to be present in wines (Jeandet et al., 1993, 1995b; Mattivi et al., 1995; Lamuela-Ravento´s and Waterhouse, 1995; Goldberg et al., 1995b,c; Soleas et al., 1995). Physiological activity of cis-resveratrol has only been described once, and was shown to inhibit kinase activity, a factor related to cancer (Jayatilake et al., 1993). There has been no specific information on the properties of cis-resveratrol, and its light sensitivity has made it difficult to purify. There is a large disparity in the reported molar absorptivities and maximum absorbance wavelengths for each of the two isomers. This has created difficulties in the quantitative reporting of individual resveratrol isomer levels. One study quantitatively expressed the cis isomer in trans-resveratrol equivalents (Lamuela-Ravento´s and Waterhouse, 1995). In other studies, a known amount of trans-resveratrol was irradiated to induce partial isomeric conversion to cis-resveratrol, and that isomerization was assumed to result in a mixture of only cis- and trans-resveratrol (Goldberg et al., 1995b; Jeandet et al., 1995b). This study determines the molar absorptivities of both compounds at two wavelengths to facilitate chromatographic quantitation at either compound’s absorbance maximum. In addition, stability trials under a variety of conditions were also conducted to test some commonly encountered laboratory conditions and their effect on the isomerization or degradation of resveratrol. © 1996 American Chemical Society

1254 J. Agric. Food Chem., Vol. 44, No. 5, 1996

Trela and Waterhouse

MATERIALS AND METHODS trans-Resveratrol was purchased from Sigma Chemical Co. (St. Louis, MO). cis-Resveratrol was obtained by irradiating a 8.94 mmol/L stock 50% ethanol solution of trans-resveratrol for 3 h. trans-Resveratrol was UV irradiated through a 2.0 mm thick borosilicate glass liquid scintillation vial placed next to a Mineralite lamp Model UVGL-25 (Ultraviolet Products, San Gabriel, CA) operating with an intensity of 180 µW/cm2 at 366 nm and, for one comparison, at 254 nm with an intensity of 750 µW/cm2. To test resveratrol stability in various buffers at differing pH values, a 50 mM phosphoric acid buffer was made using 85% phosphoric acid. Sulfuric acid was used to lower the pH, and ammonium hydroxide was used to raise buffer pH. The four buffers used were 1.0, 3.5, 7.0, and 10.0 pH units. Resveratrol solutions were mixed in a 1:1 ratio with the buffers for a final concentration of 276 µmol/L and 50% EtOH. trans-Resveratrol evaporation was conducted using rotary evaporation in vacuo at 37 °C in EtOH and the various buffers. Copper contamination (CuSO4) at the 3.2 and 7.9 µmol/L copper (Cu2+) levels and with rotary evaporation was also tested in triplicate. UV spectra were obtained with a Hewlett-Packard (HP) (Palo Alto, CA) 8452A diode array spectrophotometer controlled by HP 89532A UV-vis software (DOS) Rev. A.00.00 through a 10 mm quartz cuvette. To reduce moisture contamination of the standards and solvents, all containers were rinsed with acetone and dried with nitrogen gas. cis and trans isomer separation and collection was performed on the 8.94 mmol/L solution and the semipreparatory system described below. Each fraction of cis-resveratrol was purified with this method at least twice. Long-term storage of standards in 100% EtOH was done at -5.0 °C in sealed, light-protected borosilicate glass containers. Semipreparative HPLC was performed with a Waters 510 pump (Milford, MA), a Rheodyne U6K injector with a 5 mL fixed loop, and an HP Model 1050 UV-visible detector set at 306 nm. The column was a Waters RCM Prep Nova-Pak HR C18, 6 µm, 25 mm × 100 mm, preceded by a 25 mm × 10 mm Guard-Pak guard column of the same specifications. Elution was carried out at 20 °C using 55% aqueous methanol mobile phase at 5 mL/min flow rate. Analytical HPLC Procedure. An HP 1090 series II with a diode array UV-visible detector was controlled by chromatography and quantitation software Chemstation (DOS) Rev. 02.00. Detector light path length was 0.6 cm. All resveratrol samples were sterile filtered through 4 mm poly(tetrafluoroethylene) (PTFE) 0.45 µm pore size filters before HPLC injection. Analyses were performed using an HP ODS Hypersil, 5 µm, 2.1 mm × 100 mm stationary phase column with a 2.1 mm × 20 mm precolumn with the same phase. The mobile phase flux was 0.5 mL/min. The eluents were (A) deionized, sterile filtered H2O and (B) methanol. Injection volume was 5.0 µL. The column was thermostatically controlled to maintain a temperature of 40.0 °C. The separation was conducted using a linear gradient of 100% solvent A to 100% solvent B in 15 min, and then the column was flushed for 5 min with solvent A to re-establish the initial conditions before another sample was injected. The eluent was monitored at 306 and 286 nm, the UV optimum absorbancies of trans- and cis-resveratrol, respectively, and the UV spectra of the eluent peaks were obtained. The results of the analyses are expressed in units of moles per liter or percent cis-resveratrol of the overall area of cisand trans-resveratrol at the specified wavelength corrected for molar absorptivity ratio difference. Expression as a percent cis-resveratrol of the total chromatographed peak areas (100% cis- and trans-resveratrol) simplifies matters and negates solvent evaporation effects. RESULTS AND DISCUSSION

The analytical HPLC method was optimized for simplicity and quick, efficient separation of trans- and cis-resveratrol in simple mixtures. This method is not

Figure 3. Typical chromatogram of cis- and trans-resveratrol at 286 nm.

Figure 4. UV spectra of chromatogram in Figure 3.

adequate for analysis of resveratrol isomers in wines, which was not addressed here. Parallel analyses, however, were carried out using a resveratrol wine analysis method developed by Lamuela-Ravento´s and Waterhouse (1995), and the results were equivalent. Since all samples were filtered before HPLC injection, possible filter adsorption of resveratrol was tested. No filter adsorption loss was observed for either isomer when done in triplicate. trans-Resveratrol eluted first with a retention time of 7.3 min, followed by cisresveratrol at 8.0 min. A typical chromatogram of the isomers is illustrated in Figure 3. Peak A at retention time 7.3 min was confirmed to be trans-resveratrol since its retention time and UV spectrum were identical to those of pure trans-resveratrol. Peak B at retention time 8.0 min was formed after brief UV irradiation of the pure trans-resveratrol standard. Comparison of its 1H NMR spectrum was consistent with that of Jayatilake et al. (1993). In addition, ultraviolet spectra for each isomer (Figure 4) were consistent with those published by Goldberg et al. (1995b), who also performed GC-MS identification, and our λmax values were similar to those of Jayatilake et al. (1993), although the molar absorptivity values described by Jayatilake et al. (1993) were much lower than those graphically reported by Goldberg et al. (1995b) or the data herein (Table 1). The UV spectrum of cis-resveratrol was not, however, consistent with that of Siemann and Creasy (1992), who reported a UV λmax of 230 nm. Thus, it appears that the spectrum reported by Siemann and Creasy (1992) was not of cis-resveratrol but of some other product of UV irradiation, as described by Roggero and GarciaParrilla (1995). UV-Induced Isomerization. A solution containing 418 µmol/L of pure trans-resveratrol was irradiated for 120 min at 366 nm and monitored for isomerization and cis-resveratrol formation every 10 min. Partial conversion to cis-resveratrol was maximum at about 100 min

J. Agric. Food Chem., Vol. 44, No. 5, 1996 1255

Resveratrol Table 1. trans- and cis-Resveratrol UV λmax (EtOH) transauthor

year

Hathway and Seakins Ingham Langcake and Pryce Siemann and Creasy

1959 1976 1976 1992

Jayatilake et al.

1993

this work

Table 2. trans- and cis-Resveratrol Molar Absorptivities (EtOH)

cis-

λmax (nm)



305a

28200b

307 306 306 320 305 320 306a 308 320 306

NDc 26700 26800 25900 10400b 2200 16200 30000 29000 31800e

absorbances

λmax (nm)



285

ND

261

22800

230 290

1600 1400

286 288 286

12500d 12600 13100e

Piceid (trans-resveratrol glucoside). b Originally expressed as log ). c ND ) not determined. d ) not λmax. e UV λmax (10% EtOH) nm (). a

concn (µmol/L)

286 nm

5.52 11.04 33.13 55.61

0.0746 0.2227 0.6850 1.0186

slope () r2 a CVb

18987 0.993 18.0

5.26 10.56 31.60 52.76

288 nm

306 nm

trans-Resveratrol 0.0846 0.1493 0.2374 0.3396 0.7405 1.0481 1.1107 1.6236 20668 0.994 15.8

29859 0.998 6.7

308 nm

320 nm

0.1514 0.3416 1.0561 1.6284

0.1517 0.3336 1.0190 1.5789

29959 0.997 6.5

28972 0.998 5.1

cis-Resveratrol (Corrected Absorbances)c 0.0726 0.0726 0.0574 0.0552 0.1473 0.1466 0.1052 0.0997 0.3987 0.3985 0.2790 0.2621 0.6732 0.6728 0.4675 0.4385

slope () r2 CV

12536 1.000 5.2

12626 1.000 5.1

8715 0.999 10.3

8163 0.999 11.4

trans/cis

1.515

1.637

3.426

3.670

a

square of Pearson’s correlation coefficient. b CV, coefficient of variance (%). c See text.

Figure 5. cis-Resveratrol conversion from a 418 µmol/L solution of 100% trans-resveratrol irradiated at 366 nm at 180 µW/cm2 and at 254 nm at 750 µW/cm2 (s, 366 nm; - - -, 254 nm).

of irradiation. trans-Resveratrol was converted to 90.6% cis-resveratrol. For comparison, another 418 µmol/L solution of 100% trans-resveratrol solution was irradiated at 254 nm for 60 min (Figure 5), although the borosilicate container would dramatically attenuate the low wavelength radiation. Even after 10 h of irradiation at 254 nm, the trans isomer had converted to e63% cisresveratrol. No other products appeared in the chromatograms after irradiation at either wavelength, and 100% of the peak areas accounted for all of the initial trans-resveratrol. The isomeric ratios in the irradiated solutions were determined using the molar absorptivity ratios from chromatographic integration data at both 286 and 306 nm. The results were the same at both wavelengths. The data confirmed that under these conditions there was a quantitative transformation of the trans isomer to the cis isomer, as described by Goldberg et al. (1995b). Another solution of 8.94 mmol/L trans-resveratrol was irradiated at 366 nm for 3 h, and the resulting g80% cis-resveratrol mixture was separated to obtain the cis-resveratrol. Calibration. Calibration curves for trans-resveratrol were obtained with g0.999 (r2) linear correlation of integration response with concentration. The curves did not vary by more than 5% over 3 months, limiting the need for recalibration, though it was performed multiple times during this time period. Standards were stored at -5 °C in 100% EtOH in the dark and protected from stray light in sealed, light-proof containers. Other sets of standards were stored in light-proof, Parafilm-sealed containers in 50% aqueous ethanol, unrefrigerated. These standards also appeared to be stable, though

r2,

solvent evaporation was noticeable through the Parafilm seal as an increase in standard HPLC area, but no new peaks were observed. Standard solutions that were unprotected from light and exposed to laboratory fluorescent lighting isomerized to about 80% cis-resveratrol over 30 days. cis-Resveratrol was extremely light sensitive, and the preparation of standards, while executed in near total darkness, allowed enough light to cause slight isomerization. Purified fractions of g98% purity yielded standards of g95% purity after handling. cis-Resveratrol stored in solution in the dark at ambient temperatures in 50% EtOH remained stable for at least 35 days over the range of 5.3-52.8 µmol/L. Detection Limits and Precision. The detection limit was measured as the concentration corresponding to the lowest signal measurable above baseline with a signal-to-noise ratio of 3:1 when done in triplicate with four standards. The limits of detection for transresveratrol were 1.2 and 0.6 µmol/L at 306 and 286 nm, respectively. The limits of detection for cis-resveratrol were 0.9 and 0.6 µmol/L at 306 and 286 nm, respectively. Instrument precision was evaluated by performing 10 replicate injections of a 22.1 µmol/L transresveratrol standard. The coefficients of variation (CV) were 1.4% and 1.7% at 306 and 286 nm, respectively, indicating very high reproducibility. Molar Absorptivities. Absorptivity was determined by measuring absorbance at four concentrations and determining the slope of the line that best fit these data (Table 2). Absorptivity was UV λmax (EtOH) nm () 288 (12 600) for cis-resveratrol, CV ) 5.1%, over the range of 5.26-52.76 µmol/L. Actual concentrations were corrected for the 95% purity based on the area of the transresveratrol peak and subtracting its absorption contribution. This gave an excellent 0.99 correlation coefficient across the standards. Absorptivity in 10% ethanol was greater, UV λmax (10% EtOH) nm () 286 (13 100), possibly due to altered hydrogen bond interactions with the water present. The coefficient of variation equaled 0.4%, performed in triplicate at only one concentration. trans-Resveratrol absorptivity was done identically and appeared to be 100% pure by HPLC. Absorptivity was again based on the slope of four measurements

1256 J. Agric. Food Chem., Vol. 44, No. 5, 1996

Figure 6. trans-Resveratrol, 276 µmol/L, in pH 10.0 buffer.

Figure 7. cis-Resveratrol, 276 µmol/L, in pH 10.0 buffer.

equaling UV λmax (EtOH) nm () 308 (30 000) with a correlation coefficient g0.99 and CV ) 6.5%. This absorptivity value is higher than previously reported, likely due to rigorous protection from light. The molar absorptivity in 10% ethanol was greater, UV λmax (10% EtOH) nm () 306 (31 800), again possibly due to hydrogen bond interactions with the water. The coefficient of variation equaled 0.3%, and it, too, was performed in triplicate at one concentration. The molar absorptivities of both cis- and transresveratrol were determined at four to five wavelengths (Table 2). The higher variability of the nonmaxima values is likely due to the higher variability in the wavelength-dependent change in absorbance at these values. Stability. When protected from light, trans-resveratrol was stable for at least 42 h (CV e 1.0%) and for at least 28 days with a CV e 4.7% in buffers pH 1-7. The initial half-life for trans-resveratrol in pH 10.0 buffer was 1.6 h (Figure 6). A 418 µmol/L trans-resveratrol solution was irradiated with Mineralite (366 nm) for 2 h, becoming 90.6% cis-resveratrol, and it was then set on the counter exposed to laboratory fluorescent lighting conditions. The percent of the cis isomer dropped to 86.1% over 60 days. New samples in the various buffers were exposed to the 366 nm UV irradiation for 5 min; the equilibrium proportion at 30+ days under laboratory lighting averaged about 86% except at pH 1.0, which had about 33.7% of the cis isomer, and at pH 10.0, which had no cis-resveratrol and very little trans-resveratrol. Under identical buffer conditions, cis-resveratrol degraded at pH 10.0 with a half-life of 63.7 h (Figure 7). cis-Resveratrol was stable at pH 7.0 (CV ) 0.7%), and less stable at pH 3.5 (CV ) 2.6%); however, at pH 1.0

Trela and Waterhouse

Figure 8. cis-Resveratrol, 276 µmol/L, in pH 1.0 buffer (- - -, trans-resveratrol; s, cis-resveratrol).

isomerization to trans-resveratrol was efficientsafter 22.8 h (t1/2), 50% of the cis form had isomerized to the trans isomer (Figure 8). As noted above, the cis isomer is sensitive to light exposure; and one 95% pure 79 µmol/L cis isomer fraction in ethanol was placed on the laboratory counter in borosilicate glass exposed to the overhead fluorescent lighting and isomerized to 91% cisresveratrol over 30 days. Thus, under the laboratory fluorescent lighting used here, the equilibrium mixture was between 86% and 91% cis-resveratrol as determined over a time span of up to 2 months at various concentrations. The exact equilibrium mixture is likely dependent on the specific spectrum of light reaching the sample. Sample dry-down operations with rotary evaporation were checked for resveratrol stability, and contrary to Pezet et al. (1994) and McMurtrey et al. (1994), no degradation of trans-resveratrol was observed after evaporation of alcohol, confirmed by Jeandet even when heated at 40 °C (unpublished results) or in buffers at various pH values except for the pH 10.0 buffer. Triplicate 1 mL unbuffered solutions of 55.6 µmol/L trans-resveratrol in 10% EtOH were rotary evaporated at 37 °C and redisolved in 1 mL of 10% EtOH. The recoveries of trans-resveratrol in the reconstituted solutions were each 82%, though no new products were observed by HPLC analysis after this procedure. When 3.2 and 7.9 µmol/L Cu2+ were added and evaporated with the resveratrol, no degradation or oxidation was observed, and the coefficients of variation at both wavelengths averaged 1.2% and 3.2% for each copper level, respectively. When performed without copper, the CV averaged 1.6% at both wavelengths. Evaporation with the buffers also showed no degradation. Analysis of an 87.6 µmol/L standard solution of trans-resveratrol in pH 7.0 buffer gave an average CV of 0.8% before evaporation and 5.7% after evaporation at both wavelengths. At pH 3.5 the CV at both wavelengths averaged 6.6% before evaporation and 3.3% after evaporation. Again, no new peaks were seen in the chromatograms. On two occasions fractions obtained by preparative HPLC decomposed to many new compounds. Numerous attempts to reproduce this occurrence failed. Glassware contamination is speculated to be the cause. Storage of dry cis-resveratrol was problematic. After one 20 mg sample was dried, it isomerized 60% into trans-resveratrol over 3 months when stored in the dark at -5 °C; the cause for this reaction was not clear. Conclusions. Molar absorptivities over a range of commonly reported wavelengths were accurately determined for both cis- and trans-resveratrol. trans-Resveratrol should be handled with caution, but can be

J. Agric. Food Chem., Vol. 44, No. 5, 1996 1257

Resveratrol

handled in the laboratory without stringent lighting precautions, whereas cis-resveratrol cannot. Laboratory lighting conditions favor an equilibrium of about 91% cis-resveratrol. Low pH causes cis-resveratrol isomerization to the trans isomer, a sterically more stable form. Some literature reports of cis-resveratrol are inaccurate and were addressed here. There is a need for a quantitative kinetic study of resveratrol isomerization under well-defined light fluxes and in the presence of acid. LITERATURE CITED Bertelli, A. A. E.; Giovannini, L.; Giannessi, D.; Migliori, M.; Bernini, W.; Fregoni, M.; Bertelli, A. Antiplatelet activity of synthetic and natural resveratrol in red wine. Int. J. Tissue React. 1995, 17, 1-3. Chung, M. I.; Teng, C. M.; Cheng, K. L.; Ko, F. N.; Lin, C. N. An antiplatelet principle of Veratrum formosanum. Planta Med. 1992, 58, 274-276. Creasy, L. L.; Coffee, M. Phytoalexin production potential of grape berries. J. Am. Soc. Hortic. Sci. 1988, 113, 230-234. Frankel, E. N.; Waterhouse, A. L.; Kinsella, J. E. Inhibition of human LDL oxidation by resveratrol. Lancet 1993, 341, 1103-1104. Fritzemeier, K.-H.; Kindl, H. Coordinate induction by UV light of stilbene synthase phenylalanine ammonia-lyase and cinnamate 4-hydroxylase in leaves of Vitaceae. Planta 1981, 151, 48-52. Goldberg, D. M.; Yan, J.; Ng, E.; Diamandis, E. P.; Karumanchiri, A.; Soleas, G.; Waterhouse, A. L. Direct injection gas chromatography mass spectrometric assay for trans-resveratrol. Anal. Chem. 1994, 66, 3959-3963. Goldberg, D. M.; Yan, J.; Ng, E.; Diamandis, E. P.; Karumanchiri, A.; Soleas, G.; Waterhouse, A. L. A global survey of trans-resveratrol concentrations in commercial wines. Am. J. Enol. Vitic. 1995a, 46, 159-165. Goldberg, D. M.; Ng, E.; Karumanchiri, A.; Yan, J.; Diamandis, E. P.; Soleas, G. J. Assay of resveratrol glycosides and isomers in wine by direct-injection high-performance liquid chromatography. J. Chromatogr. A 1995b, 708, 89-98. Goldberg, D. M.; Karumanchiri, A.; Ng, E.; Yan, J.; Diamandis, E. P.; Soleas, G. J. Direct gas chromatographic-mass spectrometric method to assay cis-resveratrol in wines: preliminary survey of its concentration in commercial wines. J. Agric. Food Chem. 1995c, 43, 1245-1250. Gorham, J. The stilbenoids. In Progress in Phytochemistry; Reinhold, L., Harborne, J. B., Swain, T., Eds.; Pergamon Press: Oxford, U. K., 1980; pp 203-252. Hathway, D. E.; Seakins, J. W. T. Hydroxystilbenes of Eucalyptus wando. Biochem. J. 1959, 72, 369-374. Ingham, J. L. 3,5,4′-Trihydroxystilbene as a phytoalexin from groundnuts (Arachis hypognea). Phytochemistry 1976, 15, 1791-1793. Jayatilake, G. S.; Jayasuriya, H.; Lee, E. S.; Koonchanok, N. M.; Geahlen, R. L.; Ashendel, C. L.; McLaughlin, J. L.; Chang, C. J. Kinase inhibitors from Polygonum cuspidatum. J. Nat. Prod. 1993, 56, 1805-1810. Jeandet, P.; Bessis, R.; Gautheron, B. The production of resveratrol (3,5,4′-trihydroxystilbene) by grape berries in different developmental stages. Am. J. Enol. Vitic. 1991, 42, 41-46. Jeandet, P.; Bessis, R.; Maume, B. F.; Sbaghi, M. Analysis of resveratrol in burgundy wines. J. Wine Res. 1993, 4, 7985. Jeandet, P.; Bessis, R.; Sbaghi, M.; Meunier, P. Production of the phytoalexin resveratrol by grapes as a response to Botrytis attacks under natural conditions. J. Phytopathol. 1995a, 143, 135-139. Jeandet, P.; Bessis, R.; Maume, B. F.; Meunier, P.; Peyron, D.; Trollat, P. Effect of enological practices on the resveratrol isomer content of wine. J. Agric. Food Chem. 1995b, 43, 316-319. Jeandet, P.; Bessis, R.; Sbaghi, M.; Meunier, P.; Trollat, P. Resveratrol content of wines of different ages: relationship

with fungal disease pressure in the vineyard. Am. J. Enol. Vitic. 1995c, 46, 1-6. Kimura, Y.; Ohminami, H.; Okuda, H.; Baba, K.; Kozawa, M.; Arichi, S. Effects of stilbene components of roots of Polygonum ssp. on liver injury in peroxidized oil fed rats. Planta Med. 1983, 49, 51-54. Kimura, Y.; Okuda, H.; Arichi, S. Effects of stilbene on arachidonate metabolism in leukocytes. Biochim. Biohys. Acta 1985, 834, 275-278. Lamuela-Ravento´s, R. M.; Waterhouse, A. L. Occurrence of resveratrol in selected California wines by a new HPLC method. J. Agric. Food Chem. 1993, 41, 521-523. Lamuela-Ravento´s, R. M.; Romero-Pe´rez, A. I.; Waterhouse, A. L.; de la Torre-Boronat, M. C. Direct HPLC analysis of cis- and trans-resveratrol and piceid isomers in Spanish red Vitis vinifera wines. J. Agric. Food Chem. 1995, 43, 281283. Langcake, P.; Pryce, R. J. The production of resveratrol by Vitis vinifera and other members of the Vitaceae as a response to infection or injury. Physiol. Plant Pathol. 1976, 9, 7786. Mattivi, F. Solid phase extraction of trans-resveratrol from wines for HPLC analysis. Z. Lebensm. Unters. Forsch. 1993, 196, 522-525. Mattivi, F.; Reniero, F.; Korhammer, S. Isolation, characterization, and evolution in red wine vinification of resveratrol monomers. J. Agric. Food Chem. 1995, 43, 1820-1823. McMurtrey, K. D.; Minn, J.; Pobanz, K.; Schultz, T. P. Analysis of wines for resveratrol using direct injection high-pressure liquid chromatography with electrochemical detection. J. Agric. Food Chem. 1994, 42, 2077-2080. Nonomura, S.; Kanagawa, H.; Makimoto, A. Chemical constituents of polygonaceous plants. I. Studies on the components of Ko-jo-kon (Polygonum cuspidatum SIEB et ZUCC.). Yakugaku Zasshi 1963, 83, 988-990. Pace-Asciak, C. R.; Hahn, S. E.; Diamandis, E. P.; Soleas, G.; Goldberg, D. M., The red wine phenolics trans-resveratrol and quercetin block human platelet aggregation and eicosanoid synthesis: Implications for protection against coronary heart disease. Clin. Chim. Acta 1995, 235, 207219. Pezet, R.; Pont, V.; Cuenat, P. Method to determine resveratrol and pterostilbene in grape berries and wines using high performance liquid chromatography and highly sensitive fluorimetric detection. J. Chromatogr. 1994, 633, 191-197. Roggero, J. P.; Archie, P. Quantitative determination of resveratrol and of one of its glycosides in wines. Sci. Aliments 1994, 14, 99-107. Roggero, J. P.; Garcia-Parrilla, C. Effects of ultraviolet irradiation on resveratrol and changes in resveratrol and various of its derivatives in the skins of ripening grapes. Sci. Aliments 1995, 15, 411-422. Schoeppner, A.; Kindl, H. Purification and properties of a stilbene synthase from induced cell suspension culture of peanuts. J. Biol. Chem. 1984, 259, 6806-6811. Siemann, E. H.; Creasy, L. L. Concentration of the phytoalexin resveratrol in wine. Am. J. Enol. Vitic. 1992, 43, 49-52. Soleas, G. J.; Goldberg, D. M.; Diamandis, E. P.; Karumanchiri, A.; Yan, J.; Ng, E. A derivatized gas chromatographic-mass spectrometric method for the analysis of both isomers of resveratrol in juice and wine. Am. J. Enol. Vitic. 1995, 46, 346-352. Vrhovsek, U.; Eder, R.; Wendelin, S. The occurrence of transresveratrol in Slovenian red and white wines. Acta Aliment. 1995, 24, 203-212. Received for review July 18, 1995. Revised manuscript received February 20, 1996. Accepted March 1, 1996.X We thank the American Vineyard Foundation and the Melini Fellowship from Wildman and Sons Inc. for their gracious support of this research. JF9504576

Abstract published in Advance ACS Abstracts, April 15, 1996. X